检索结果-内蒙古大学图书馆

30th IEEE international symposium on high-performance computer architecture (HPCA)

作者： Li, Yueying Lazarev, Nikita Koufaty, David Yin, Tenny Anderson, Andy Zhang, Zhiru Suh, G. Edward Kaffes, Kostis Delimitrou, Christina Cornell Univ Ithaca NY 14853 USA MIT Cambridge MA USA Intel Labs Hillsboro OR USA Columbia Univ New York NY USA

ISBN: (纸本)9798350393132;9798350393149

Modern cloud applications are prone to high tail latencies since their requests typically follow highly-dispersive distributions. Prior work has proposed both OS- and systemlevel solutions to reduce tail latencies for microsecond-scale workloads through better scheduling. Unfortunately, existing approaches like customized dataplane OSes, require significant OS changes, experience scalability limitations, or do not reach the full performance capabilities hardware offers. We propose LibPreemptible, a preemptive user-level threading library that is flexible, lightweight, and scalable. LibPreemptible is based on three key techniques: 1) a fast and lightweight hardware mechanism for delivery of timed interrupts, 2) a general-purpose user-level scheduling interface, and 3) an API for users to express adaptive scheduling policies tailored to the needs of their applications. Compared to the prior state-of-the-art scheduling system Shinjuku, our system achieves significant tail latency and throughput improvements for various workloads without the need to modify the kernel. We also demonstrate the flexibility of LibPreemptible across scheduling policies for real applications experiencing varying load levels and characteristics.

关键词： cloud computing datacenter performance and quality of service scheduling

来源：评论

学校读者我要写书评

暂无评论

Peer-to-peer checkpointing arrangement for mobile grid computing systems 07

Peer-to-peer checkpointing arrangement for mobile grid compu...

引用

16th international symposium on high performance Distributed computing 2007, HPDC'07 and Co-Located workshops

作者： Darby III, Paul J. Tzeng, Nian Feng Center for Advanced Computer Studies University of Louisiana at Lafayette Lafayette LA 70504

ISBN: (纸本)1595936734

This paper deals with a novel, distributed, QoS-aware, peer-to-peer checkpointing arrangement component for Mobile Grid (MoG) computing systems middleware. Checkpointing is more crucial in MoG systems than in their wired counterparts due to node mobility and less reliable wireless links resulting in frequent and dynamic connections and disconnections. Having determined the globally optimal checkpoint arrangement to be NP-complete, we consider ReD, our Reliability Driven (ReD) protocol, employing QoS-aware heurisitcs, for constucting superior peer-to-peer checkpointing arrangements efficiently.

关键词： Distributed computer systems

来源：评论

学校读者我要写书评

暂无评论

Balancing load of GPU subsystems to accelerate image reconstruction in parallel beam tomography 30

Balancing load of GPU subsystems to accelerate image reconst...

引用

30th international symposium on computer architecture and high performance computing (SBAC-PAD)

作者： Chilingaryan, Suren Ametova, Evelina Kopmann, Andreas Mirone, Alessandro Karlsruhe Inst Technol Karlsruhe Germany Katholieke Univ Leuven Leuven Belgium ESRF Grenoble France

ISBN: (纸本)9781538677698

Synchrotron X-ray imaging is a powerful method to investigate internal structures down to the micro-and nanoscopic scale. Fast cameras recording thousands of frames per second allow time-resolved studies with a high temporal resolution. Fast image reconstruction is essential to provide the synchrotron instrumentation with the imaging information required to track and control the process under study. Traditionally Filtered Back Projection algorithm is used for tomographic reconstruction. In this article, we discuss how to implement the algorithm on nowadays GPGPU architectures efficiently. The key is to achieve balanced utilization of available GPU subsystems. We present two highly optimized algorithms to perform back projection on parallel hardware. One is relying on the texture engine to perform reconstruction, while another one utilizes the Core computational units of the GPU. Both methods outperform current state-of-the-art techniques found in the standard reconstructions codes significantly. Finally, we propose a hybrid approach combining both algorithms to better balance load between GPU subsystems. It further boosts the performance by about 30% on NVIDIA Pascal micro-architecture.

关键词： Graphics processing units Image reconstruction Engines computer architecture Open area test sites Tomography

来源：评论

学校读者我要写书评

暂无评论

Many SVDs on GPU for Image Mosaic Assemble 27

Many SVDs on GPU for Image Mosaic Assemble

引用

IEEE 27th international symposium on computer architecture and high performance computing

作者： Badolato, Irving de Paula, Luciano Farias, Ricardo Univ Fed Rio de Janeiro COPPE Syst Engn Ctr Technol Bldg HRoom 319 BR-21945970 Rio De Janeiro RJ Brazil

ISBN: (纸本)9781467386210

In this paper we present a homography algorithm to produce image mosaics using parallelism to solve a multiple Singular Value Decomposition (SVD) system. We analyse four state of art SVD methods and choose the one which better suites the expected size of the matrices derived from the datasets of interest. Then we use cuda to accelerate the solution of the transformation homogeneous matrices.

关键词： Homography GPU computing SVD solver

来源：评论

学校读者我要写书评

暂无评论

Phase-Based Tasks Scheduling in Data Centers Powered Exclusively by Renewable Energy 31

Phase-Based Tasks Scheduling in Data Centers Powered Exclusi...

引用

31st international symposium on computer architecture and high performance computing (SBAC-PAD)

作者： Caux, Stephane Renaud-Goud, Paul Rostirolla, Gustavo Stolf, Patricia Univ Toulouse IRIT F-31062 Toulouse France Univ Toulouse Laplace UMR5213 INPT UPSCNRS Toulouse France

ISBN: (纸本)9781728141947

Data centers are considered nowadays as the factories of the digital age, being currently responsible for consuming more energy than the entire United Kingdom. On the other side, the global total capacity of renewable power increases continuously. The combination of these two factors calls for new approaches in designing data centers powered only by renewable energy sources. Our work focuses on task scheduling optimization under a power envelope, and on the way to handle power starvation, i.e. when the available power does not provide sufficient resources to execute a given workload. To do so we utilize the concept of task degradation through cross-correlation to find where to place the tasks in order to reduce the data center profit degradation. The results show that our algorithm could obtain more than 34% increase in profit when compared to algorithms from the literature, while fulfilling the power profile and resources constraints.

关键词： Scheduling Cloud computing Renewable Energy Mixed Workload

来源：评论

学校读者我要写书评

暂无评论

Introduction to the special issue on the 18th international symposium on computer architecture and high performance computing

引用

international JOURNAL OF PARALLEL PROGRAMMING 2008年第2期36卷 163-165页

作者： De Souza, Alberto F. Buyya, Rajkumar Univ Fed Espirito Santo Dept Informat Lab Comp Alto Desemmpenho Vitoria ES Brazil Univ Melbourne Dept Comp Sci & Software Engn Grid Comp & Distributed Syst Lab Carlton Vic Australia

来源：评论

学校读者我要写书评

暂无评论

Automatic Ray-Tracer Cloud Offloading in OpenMP 30

Automatic Ray-Tracer Cloud Offloading in OpenMP

引用

30th international symposium on computer architecture and high performance computing (SBAC-PAD)

作者： Mortatti, Matheus Yviquel, Herve Araujo, Guido Univ Campinas UNICAMP Inst Comp Campinas SP Brazil

ISBN: (纸本)9781538677698

Rendering an image from a 3D scene requires a large amount of computation which grows exponentially with the complexity of the scene (e.g. number of objects and light sources). With the increasing demand of high definition content, 3D designers need to use high-performance computer systems to keep the rendering time acceptable. Since owning computer clusters is expensive, designers usually rent computing power directly from cloud service providers (e.g. AWS and Azure). However, even though many cloud providers already propose dedicated rendering services, integrating them within the standard workflow of modeling softwares can become a complex and cumbersome task. It typically requires exporting the project from the design software, dealing with various access control mechanisms from different clouds to upload the project, and executing the rendering remotely through command-line. Offloading computation to the cloud is a technique which can considerably simplify such tasks. To achieve that, this paper uses an extension of OpenMP 4. X to eliminate any major interactions with the end-user, while minimizing the complexity of cloud integration and optimizing the design workflow. It applies such approach to a ray-tracing application, a simplified version of the engines used by professional 3D modeling software (e.g. Blender). It automatically offloads the rendering process from the user computer to computer cluster within the Microsoft Azure cloud, brings the resulting images back after the computation ends and displays them directly on the screen of the user computer, thus providing a transparent programming model and good speed-ups over local execution.

关键词： Cloud computing Rendering (computer graphics) Computational modeling Software Three-dimensional displays Cameras Engines

来源：评论

学校读者我要写书评

暂无评论

high performance and Portable Convolution Operators for Multicore Processors 32

High Performance and Portable Convolution Operators for Mult...

引用

32nd IEEE international symposium on computer architecture and high-performance computing (SBAC-PAD) / 11th Workshop on Applications for Multi-Core architectures (WAMCA)

作者： San Juan, Pablo Castello, Adrian Dolz, Manuel F. Alonso-Jorda, Pedro Quintana-Orti, Enrique S. Univ Politecn Valencia Valencia Spain Univ Jaume 1 Castellon de La Plana Spain

ISBN: (纸本)9781728199245

The considerable impact of Convolutional Neural Networks on many Artificial Intelligence tasks has led to the development of various high performance algorithms for the convolution operator present in this type of networks. One of these approaches leverages the IM2COL transform followed by a general matrix multiplication (GEMM) in order to take advantage of the highly optimized realizations of the GEMM kernel in many linear algebra libraries. The main problems of this approach are 1) the large memory workspace required to host the intermediate matrices generated by the IM2COL transform;and 2) the time to perform the IM2COL transform, which is not negligible for complex neural networks. This paper presents a portable high performance convolution algorithm based on the BLIS realization of the GEMM kernel that avoids the use of the intermediate memory by taking advantage of the BLIS structure. In addition, the proposed algorithm eliminates the cost of the explicit IM2COL transform, while maintaining the portability and performance of the underlying realization of GEMM in BLIS.

关键词： Convolutional neural networks high performance multicore processors

来源：评论

学校读者我要写书评

暂无评论

SpecFaaS: Accelerating Serverless Applications with Speculative Function Execution 29

SpecFaaS: Accelerating Serverless Applications with Speculat...

引用

29th IEEE international symposium on high-performance computer architecture (HPCA)

作者： Stojkovic, Jovan Xu, Tianyin Franke, Hubertus Torrellas, Josep Univ Illinois Urbana IL 61801 USA IBM Res Yorktown Hts NY USA

ISBN: (纸本)9781665476522

Serverless computing has emerged as a popular cloud computing paradigm. Serverless environments are convenient to users and efficient for cloud providers. However, they can induce substantial application execution overheads, especially in applications with many functions. In this paper, we propose to accelerate serverless applications with a novel approach based on software-supported speculative execution of functions. Our proposal is termed Speculative Function-as-a-Service (SpecFaaS). It is inspired by out-of-order execution in modern processors, and is grounded in a characterization analysis of FaaS applications. In SpecFaaS, functions in an application are executed early, speculatively, before their control and data dependences are resolved. Control dependences are predicted like in pipeline branch prediction, and data dependences are speculatively satisfied with memoization. With this support, the execution of downstream functions is overlapped with that of upstream functions, substantially reducing the end-to-end execution time of applications. We prototype SpecFaaS on Apache OpenWhisk, an open-source serverless computing platform. For a set of applications in a warmed-up environment, SpecFaaS attains an average speedup of 4.6x. Further, on average, the application throughput increases by 3.9x and the tail latency decreases by 58.7%.

关键词： Cloud computing Serverless computing Function-as-a-Service

来源：评论

学校读者我要写书评

暂无评论

Customizable Heterogeneous Acceleration for Tomorrow's high-performance computing 17

Customizable Heterogeneous Acceleration for Tomorrow's High-...

引用

2015 IEEE 17th international Conference on high performance computing and Communications (HPCC)

作者： Cilardo, Alessandro Flich, Jose Gagliardi, Mirko Gavila, Rafael T. Univ Naples Federico II DIETI Naples Italy Univ Politecn Valencia GAP Valencia Spain

ISBN: (纸本)9781479989379

high-performance computing as we know it today is experiencing unprecedented changes, encompassing all levels from technology to use cases. This paper explores the adoption of customizable, deeply heterogeneous manycore systems for future QoS-sensitive and power-efficient high-performance computing. At the heart of the proposed architecture is a NoC-based manycore system embracing medium-end CPUs, GPU-like processors, and reconfigurable hardware regions. The paper discusses the high-level design principles inspiring this innovative architecture as well as the key role that heterogeneous acceleration, ranging from multicore processors and GPUs down to FPGAs, might play for tomorrow's high-performance computing.

关键词： field programmable gate arrays graphics processing units multiprocessing systems network-on-chip parallel processing power aware computing quality of service FPGA GPU-like processors NoC-based many-core system QoS-sensitive computing customizable heterogeneous acceleration heterogeneous acceleration heterogeneous manycore systems high-level design principles high-performance computing innovative architecture medium-end CPU multicore processors power-efficient high-performance computing reconfigurable hardware regions Acceleration computer architecture Field programmable gate arrays Hardware Program processors Quality of service Registers Field programmable gate arrays high performance computing Quality of service power aware computing Program processors Multiprocessing systems network-on-chip computer architecture Registers acceleration PARALLEL PROCESSING (computerS) Multi-core processors computer hardware Graphics Processing Unit

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：